IMHO, being able to run your program on *any* modern computer is a huge boon. For starters, it gives you a more advance way to interact with a computer, which drives home that you aren't just playing around with a form of abstract math. From there, you get to use it in your everyday life (e.g. auto-fill a survey with a bookmarklet) and show off your skills to your friends.
It's actually rather common for new drugs to be tested on the terminally ill, for exactly the reasons you state. The thing is, though, they actually have a chance of working, hence why it's being researched. A lot of stem cell companies promise miracles to people at their most vulnerable and charge exorbitant fees for snake oil.
Informed consent is not compatible with a profit incentive unless you have a proven treatment. I don't know if Celltex is like other stem cell companies, but most of them will spread disinformation about outright curing hundreds of diseases and downplay the risks to a mere "we're not liable for anything" clause.
OTOH, it enables them to continue in their current state. Hopefully that state is improved with the new regime, but if it's not then we've just prolonged it.
If cells could divide 50 times without their less fit offspring dying, a single bacterium could produce one kilogram of offspring in about a day. For perspective, in two more days its offspring would weigh much more than the Earth. Plus, most mutations are silent, or inconsequential, such as the ones that make-up your fingerprint.
Telomerase is a protein. It has a gene, but it's not a repeating segment (the promoter/enhancer might be, but that's not what you mean). Telomeres are repeating segments that fold-over onto themselves to protect chromosome ends. (Folding might just be in yeast, I forget...)
As for divvying-up work, without telomeres your most eager stem cell would likely divide thousands of times, while the others just sat around. This would almost guarantee cancer, so each cell is limited, thus requiring all stem cells to do some of the work. Each cell has a threshold for taking action, leading to some being more eager that others, so this is a common problem in the body.
The number is about eight for common cancers. Generally you need about three or so before the mutation rate increases dramatically. Still, that works out to about a one in ten thousand chance of getting those three mutations in a cell, in highly active cells (~30 divisions) which are fairly rare in end-organ tissue where cancers arise. While the human body has ~10 trillion cells, 90% of those are symbiotic bacteria, and breast/prostate/lung tissue isn't very dense. From there, the immune system kills the vast majority of cancers, so it seems like the mutation rate ought to be higher to explain cancer's prevalence. Environmental exposure can explain a lot, but low level damage can easily be controlled by the various DNA repair mechanisms (unless you are BRCA positive or something).
Telomeres are a limiter that is easily overcome by telomerase. What they are a limiter for is generally thought to be the overall number of mutations in the chromosome (which is important for multicellular organisms, so a cell doesn't evolve autonomy -- cancer). Chromosome ends aren't particularly worthy of protection themselves, as you can do just fine with a circular chromosome that lacks them all-together.
Sorry, it appears the rate was revised the year I graduated, so I was using older information (1 per 600,000). I will admit that one error per three genes from replication alone did seem too high to me. In vivo error rates seem to be one per 10^9 base pairs. Given that it's a review article, I'd have to do a lot of reading to determine how DNA packing and such affect that rate (or how they measured in vivo rates rather that ideal in vitro).
That rate would only allow for 150 mutations per cell before hitting its telemerase limit (which most do not reach). Given the number of genes, number of cells, and sequence required for cancer to form, this number seems much too low. Thus, the environmental mutation rate must make-up the difference. For what I posted, it's not terribly relevant if the mutation was replication-induced or mutagen-induced, so I essentially conflated them for simplicity. Each organism will have very different rates, so deriving highly accurate numbers isn't necessary for explaining the general concept of the purpose of teleomeres.
Offspring for each generation are subjected to natural selection, which can be thought of as an independent correcting factor in this case. A sperm or egg with detrimental mutations won't ever reach its counterpart. Most people are familiar with how many millions of sperm don't make it, but for eggs each woman has ~400,000 and only ~400 of those activate (the healthiest tend to do so earlier in life). At least 25% of fertilized eggs self-abort in the first trimester. From there, the number of fetuses that go on to reproduce is substantially less than 100%.
If you're looking for parking in Tianamen square then how is Google supposed to know what you want when you just type "Tianamen square"? Isn't it worse if they have so much data on you they can predict what you're searching for without you even needing to type it?
In humans, telomeres limit cells to ~50 divisions, which is probably related to how DNA replication is only 99.9998% accurate. After that many divisions, the genome is 0.001% different from when it started, which is one error per 10,000 base pairs, or an error in 1/3 of all genes. This is in addition to the slow rate of spontaneous mutations you accumulate over your lifetime.
In general, fatal mutations don't matter, the stem cell will just divide again (or be dead), and cells are specialized so only a small number of genes are relevant. Furthermore, cells work together, so if two nearby cells have different lineages then they have different errors, and can likely compensate for each other. Still, you don't want too many errors in your cell replication control genes (i.e. protooncogenes ==> cancer), nor can cells function well with a tremendous number of errors (i.e. "aging"). Telomeres also help divvy-up the workload among stem cells so the most eager doesn't monopolize the work.
For flatworms, all this likely entails a fast mutation rate. So what if 90% of its offspring die? The one that takes hold in a new host can produce thousands of offspring, and quickly changing their immunologic profile increases the odds of that.
The sample also seems to omit prisoners, which are more likely to be poor, and more likely to be amoral, so that biases the poor end of the sample towards ethical behavior. Beyond that, the sampling methodology (internet access or Berkeley undergraduates) is terrible.
First and foremost, both of those factors are associated with affluence, so you're biased in your sampling to start. Second, Berkeley students aren't representative of the population at large (as has been explained), nor are internet-users. For all we know, the internet users are all people who don't block advertisements, pirate everything, and cheat in online games.
Are you including all the preventative measures that are generally taken against lowlifes? E.g. theft insurance, alarm systems, locks, handguns, guard dogs, fences, security cameras, security lights, RFID tags, security personnel, and all the industries that have grown-up around each of these items.
I would expect all these measures to be rather expensive, especially when added to the losses that occur when it isn't enough. I'll not get into the conversion rate for when some lowlife decides to be a robber rather than a burglar.
A flaw. Personally, I used to allow Facebook access to my contacts because it's useful to have all my Facebook contacts synced with my phone. Only later did I learn that they upload your phone's whole address book and, by extension, your Google address book to Facebook.
BTW, if you don't think you have a Facebook account, try to think if someone who uses Facebook has entered information about you into their phone book... I know I received a suggestion to friend someone because I used to have their contact information in my phone a year ago (deleted shortly thereafter), and they joined Facebook about a month ago (no other contact or friends in common).
One problem is that many impoverished people life in what has been called a "fast food desert". IOW, the ratio of restaurants to grocery stores is ridiculously high. Financially, that means that trips to the grocery store cost gas, time, and must be infrequent. The latter stipulation means one needs to buy in bulk, which requires you have $100 or so saved up (serious issue if you live paycheck to paycheck).
As for cost, I can easily get 440 Calorie McDonalds double cheeseburgers for $1 each in less than five minutes. That's quite a bit cheaper than your proposed $5-6 meal per Calorie or per pound and requires far less time, preparation, equipment, and skill. Furthermore, your palette becomes accustomed to fast food. This serves to curtail many attempts to "learn how to cook", because everything you make tastes terrible for two reasons: you suck at cooking and you're used to fast food.
Obviously, none of this in an insurmountable problem. But it should be equally obvious that it's not that easy to escape the effects of poverty. Yes, it wastes money to eat fast food all the time (overall), but if poor people didn't waste money (or not work, or have too many dependents) then they wouldn't stay poor.
IMHO, it'd be easier to pay $5/month for a private VPN service (over port 443) and forgo paying for internet service twice. But I wholeheartedly agree with this approach. Based on how the network is setup, it could well be trivial for someone to sniff your internet traffic.
An added benefit is that nobody is likely to hassle you unless there's a subpoena, so you won't have to explain why you visited "some Chinese gaming site" to a local authoritarian (it's a wonder why I'm not even more paranoid having had such noisy network administrators).
A human needs a prescription for most of these antibiotics, in part due to side effects, but also to slow down bacterial resistance. I'm still shocked that an animal doesn't require a similar prescription from a veterinarian for exactly the same reasons. It's not a free market if someone has the law specifically made in their favor...
The blood brain barrier refers to the tight junctions between endothelial cells in the capillaries of the brain. With age these junctions loosen. Here is a scanning electron microscope picture of such a capillary.
A stroke generally involves a macroscopic embolus getting stuck in an artery in the brain. As-in a pathologist can often physically find it during an autopsy (I once heard of one that showed how one fit together with a thrombus in the leg much like two pieces of a jigsaw puzzle). The scale of the BBB and a thrombus is completely different.
IOW, a stroke involves blocking an artery, which happens long before it reaches the capillaries and BBB. You can contrast this with a pulmonary embolus which is usually a shower of small clots blocking smaller blood vessels in the lungs (although you can get things like a saddle embolus where a large clot blocks both pulmonary arteries... it's very bad).
If you want the experience, just buy an airline ticket with 200 lbs of pennies ($360). Although I'm not sure if that's more the experience of a rich person, or someone from Zimbabwe.
I used to just change my shell to Litestep and my keyboard to Dvorak then laugh as people fumbled. It's also was a kick to watch an "I'll install it myself"-type tech scratch his head when C:\Program Files and C:\Documents and Settings had both been moved to dedicated partitions, and C:\ was 1 GB in capacity.
It seems like a good solution would be to use one of those commercial "Private Internet Access" type VPNs, and firewall the VM such that it has access to no network adapter except the virtual one. The VM gets unfettered internet access and your local network is safe (so long as nothing breaks out of the VM). IMHO, it'd be crazy to not have such a VPN if you have kid approaching puberty, given how their curiosity might be difficult for Big Brother to distinguish from your perversion.
IMHO, it's more like if we had 30 MPG cars with 20 gallon fuel tanks back then, and now we had 20,000 gallon fuel tanks and cars ran at 0.04 MPG. Efficiency has gone way down as hardware has gotten about three orders of magnitude faster. Sure, in this hypothetical example range is a little better, and engineers wouldn't need to spend as much time tweaking for performance, but... wow...
With a safe, if it's locked, the contents still exist.
And that is why I always place my important documents in a locked safe with a tiny radiation source and a Geiger counter. If the Geiger counter detects radiation, then a thermite charge is activated. Due to my poor understanding of a 77 year old reductio ad absurdum of the Copenhagen interpretation, my documents neither exist nor don't exist! And I surely cannot be compelled to collapse the waveform by a court of law, the constitution gives them no power over quantum physics.
On the Android market, a terrible update (i.e. the last twenty consecutive reviews are one star) may cause the app's rating to drop half a star over a few days. For historically good apps, the developers can get away with quite a bit of user abuse before their ratings suffer noticeably.
IMHO, Google should integrate the change log and post-update reviews into the install screen for any update. Also, ratings of the most recent version of the app should have far more weight (obviously factoring in people that hate change, developers that may release rapid updates to game the system, and developers that abandon their app so it no longer functions).
Suspension of disbelief has limits. Movie writers seem to rely on mass ignorance to stretch these limits. Nobody would accept a car chase that suddenly goes airborne because everybody knows that cars can't fly under their own power. That said, some things can be accepted as a visually appealing metaphor, such as banging at the keyboard to hack a computer.
Others, like being knocked out for half a day and actually waking up (and without brain damage no less), simply shows the writer knows nothing about the subject he's writing on. Once you realize the world doesn't follow the same laws as your own, and there is no predictable set of laws that it does follow, then most people lose interest. The semi-predictability is crucial. It's akin to the difference between a conversation and a random series of words. Or a protagonist which has 30 superpowers and solves every conflict by revealing a new one (deus ex machina).
Slashdot Mirror
IMHO, being able to run your program on *any* modern computer is a huge boon. For starters, it gives you a more advance way to interact with a computer, which drives home that you aren't just playing around with a form of abstract math. From there, you get to use it in your everyday life (e.g. auto-fill a survey with a bookmarklet) and show off your skills to your friends.
It's actually rather common for new drugs to be tested on the terminally ill, for exactly the reasons you state. The thing is, though, they actually have a chance of working, hence why it's being researched. A lot of stem cell companies promise miracles to people at their most vulnerable and charge exorbitant fees for snake oil.
Informed consent is not compatible with a profit incentive unless you have a proven treatment. I don't know if Celltex is like other stem cell companies, but most of them will spread disinformation about outright curing hundreds of diseases and downplay the risks to a mere "we're not liable for anything" clause.
OTOH, it enables them to continue in their current state. Hopefully that state is improved with the new regime, but if it's not then we've just prolonged it.
If cells could divide 50 times without their less fit offspring dying, a single bacterium could produce one kilogram of offspring in about a day. For perspective, in two more days its offspring would weigh much more than the Earth. Plus, most mutations are silent, or inconsequential, such as the ones that make-up your fingerprint.
Telomerase is a protein. It has a gene, but it's not a repeating segment (the promoter/enhancer might be, but that's not what you mean). Telomeres are repeating segments that fold-over onto themselves to protect chromosome ends. (Folding might just be in yeast, I forget...)
As for divvying-up work, without telomeres your most eager stem cell would likely divide thousands of times, while the others just sat around. This would almost guarantee cancer, so each cell is limited, thus requiring all stem cells to do some of the work. Each cell has a threshold for taking action, leading to some being more eager that others, so this is a common problem in the body.
The number is about eight for common cancers. Generally you need about three or so before the mutation rate increases dramatically. Still, that works out to about a one in ten thousand chance of getting those three mutations in a cell, in highly active cells (~30 divisions) which are fairly rare in end-organ tissue where cancers arise. While the human body has ~10 trillion cells, 90% of those are symbiotic bacteria, and breast/prostate/lung tissue isn't very dense. From there, the immune system kills the vast majority of cancers, so it seems like the mutation rate ought to be higher to explain cancer's prevalence. Environmental exposure can explain a lot, but low level damage can easily be controlled by the various DNA repair mechanisms (unless you are BRCA positive or something).
Telomeres are a limiter that is easily overcome by telomerase. What they are a limiter for is generally thought to be the overall number of mutations in the chromosome (which is important for multicellular organisms, so a cell doesn't evolve autonomy -- cancer). Chromosome ends aren't particularly worthy of protection themselves, as you can do just fine with a circular chromosome that lacks them all-together.
Sorry, it appears the rate was revised the year I graduated, so I was using older information (1 per 600,000). I will admit that one error per three genes from replication alone did seem too high to me. In vivo error rates seem to be one per 10^9 base pairs. Given that it's a review article, I'd have to do a lot of reading to determine how DNA packing and such affect that rate (or how they measured in vivo rates rather that ideal in vitro).
That rate would only allow for 150 mutations per cell before hitting its telemerase limit (which most do not reach). Given the number of genes, number of cells, and sequence required for cancer to form, this number seems much too low. Thus, the environmental mutation rate must make-up the difference. For what I posted, it's not terribly relevant if the mutation was replication-induced or mutagen-induced, so I essentially conflated them for simplicity. Each organism will have very different rates, so deriving highly accurate numbers isn't necessary for explaining the general concept of the purpose of teleomeres.
Offspring for each generation are subjected to natural selection, which can be thought of as an independent correcting factor in this case. A sperm or egg with detrimental mutations won't ever reach its counterpart. Most people are familiar with how many millions of sperm don't make it, but for eggs each woman has ~400,000 and only ~400 of those activate (the healthiest tend to do so earlier in life). At least 25% of fertilized eggs self-abort in the first trimester. From there, the number of fetuses that go on to reproduce is substantially less than 100%.
If you're looking for parking in Tianamen square then how is Google supposed to know what you want when you just type "Tianamen square"? Isn't it worse if they have so much data on you they can predict what you're searching for without you even needing to type it?
In humans, telomeres limit cells to ~50 divisions, which is probably related to how DNA replication is only 99.9998% accurate. After that many divisions, the genome is 0.001% different from when it started, which is one error per 10,000 base pairs, or an error in 1/3 of all genes. This is in addition to the slow rate of spontaneous mutations you accumulate over your lifetime.
In general, fatal mutations don't matter, the stem cell will just divide again (or be dead), and cells are specialized so only a small number of genes are relevant. Furthermore, cells work together, so if two nearby cells have different lineages then they have different errors, and can likely compensate for each other. Still, you don't want too many errors in your cell replication control genes (i.e. protooncogenes ==> cancer), nor can cells function well with a tremendous number of errors (i.e. "aging"). Telomeres also help divvy-up the workload among stem cells so the most eager doesn't monopolize the work.
For flatworms, all this likely entails a fast mutation rate. So what if 90% of its offspring die? The one that takes hold in a new host can produce thousands of offspring, and quickly changing their immunologic profile increases the odds of that.
The sample also seems to omit prisoners, which are more likely to be poor, and more likely to be amoral, so that biases the poor end of the sample towards ethical behavior. Beyond that, the sampling methodology (internet access or Berkeley undergraduates) is terrible.
First and foremost, both of those factors are associated with affluence, so you're biased in your sampling to start. Second, Berkeley students aren't representative of the population at large (as has been explained), nor are internet-users. For all we know, the internet users are all people who don't block advertisements, pirate everything, and cheat in online games.
Are you including all the preventative measures that are generally taken against lowlifes? E.g. theft insurance, alarm systems, locks, handguns, guard dogs, fences, security cameras, security lights, RFID tags, security personnel, and all the industries that have grown-up around each of these items.
I would expect all these measures to be rather expensive, especially when added to the losses that occur when it isn't enough. I'll not get into the conversion rate for when some lowlife decides to be a robber rather than a burglar.
A flaw. Personally, I used to allow Facebook access to my contacts because it's useful to have all my Facebook contacts synced with my phone. Only later did I learn that they upload your phone's whole address book and, by extension, your Google address book to Facebook.
BTW, if you don't think you have a Facebook account, try to think if someone who uses Facebook has entered information about you into their phone book... I know I received a suggestion to friend someone because I used to have their contact information in my phone a year ago (deleted shortly thereafter), and they joined Facebook about a month ago (no other contact or friends in common).
One problem is that many impoverished people life in what has been called a "fast food desert". IOW, the ratio of restaurants to grocery stores is ridiculously high. Financially, that means that trips to the grocery store cost gas, time, and must be infrequent. The latter stipulation means one needs to buy in bulk, which requires you have $100 or so saved up (serious issue if you live paycheck to paycheck).
As for cost, I can easily get 440 Calorie McDonalds double cheeseburgers for $1 each in less than five minutes. That's quite a bit cheaper than your proposed $5-6 meal per Calorie or per pound and requires far less time, preparation, equipment, and skill. Furthermore, your palette becomes accustomed to fast food. This serves to curtail many attempts to "learn how to cook", because everything you make tastes terrible for two reasons: you suck at cooking and you're used to fast food.
Obviously, none of this in an insurmountable problem. But it should be equally obvious that it's not that easy to escape the effects of poverty. Yes, it wastes money to eat fast food all the time (overall), but if poor people didn't waste money (or not work, or have too many dependents) then they wouldn't stay poor.
IMHO, it'd be easier to pay $5/month for a private VPN service (over port 443) and forgo paying for internet service twice. But I wholeheartedly agree with this approach. Based on how the network is setup, it could well be trivial for someone to sniff your internet traffic.
An added benefit is that nobody is likely to hassle you unless there's a subpoena, so you won't have to explain why you visited "some Chinese gaming site" to a local authoritarian (it's a wonder why I'm not even more paranoid having had such noisy network administrators).
I trust you take a consistent approach on non-degree-relevant telephone calls, television channels, and postal service in the dorms?
A human needs a prescription for most of these antibiotics, in part due to side effects, but also to slow down bacterial resistance. I'm still shocked that an animal doesn't require a similar prescription from a veterinarian for exactly the same reasons. It's not a free market if someone has the law specifically made in their favor...
The blood brain barrier refers to the tight junctions between endothelial cells in the capillaries of the brain. With age these junctions loosen. Here is a scanning electron microscope picture of such a capillary.
A stroke generally involves a macroscopic embolus getting stuck in an artery in the brain. As-in a pathologist can often physically find it during an autopsy (I once heard of one that showed how one fit together with a thrombus in the leg much like two pieces of a jigsaw puzzle). The scale of the BBB and a thrombus is completely different.
IOW, a stroke involves blocking an artery, which happens long before it reaches the capillaries and BBB. You can contrast this with a pulmonary embolus which is usually a shower of small clots blocking smaller blood vessels in the lungs (although you can get things like a saddle embolus where a large clot blocks both pulmonary arteries... it's very bad).
If you want the experience, just buy an airline ticket with 200 lbs of pennies ($360). Although I'm not sure if that's more the experience of a rich person, or someone from Zimbabwe.
I used to just change my shell to Litestep and my keyboard to Dvorak then laugh as people fumbled. It's also was a kick to watch an "I'll install it myself"-type tech scratch his head when C:\Program Files and C:\Documents and Settings had both been moved to dedicated partitions, and C:\ was 1 GB in capacity.
It seems like a good solution would be to use one of those commercial "Private Internet Access" type VPNs, and firewall the VM such that it has access to no network adapter except the virtual one. The VM gets unfettered internet access and your local network is safe (so long as nothing breaks out of the VM). IMHO, it'd be crazy to not have such a VPN if you have kid approaching puberty, given how their curiosity might be difficult for Big Brother to distinguish from your perversion.
IMHO, it's more like if we had 30 MPG cars with 20 gallon fuel tanks back then, and now we had 20,000 gallon fuel tanks and cars ran at 0.04 MPG. Efficiency has gone way down as hardware has gotten about three orders of magnitude faster. Sure, in this hypothetical example range is a little better, and engineers wouldn't need to spend as much time tweaking for performance, but... wow...
With a safe, if it's locked, the contents still exist.
And that is why I always place my important documents in a locked safe with a tiny radiation source and a Geiger counter. If the Geiger counter detects radiation, then a thermite charge is activated. Due to my poor understanding of a 77 year old reductio ad absurdum of the Copenhagen interpretation, my documents neither exist nor don't exist! And I surely cannot be compelled to collapse the waveform by a court of law, the constitution gives them no power over quantum physics.
On the Android market, a terrible update (i.e. the last twenty consecutive reviews are one star) may cause the app's rating to drop half a star over a few days. For historically good apps, the developers can get away with quite a bit of user abuse before their ratings suffer noticeably.
IMHO, Google should integrate the change log and post-update reviews into the install screen for any update. Also, ratings of the most recent version of the app should have far more weight (obviously factoring in people that hate change, developers that may release rapid updates to game the system, and developers that abandon their app so it no longer functions).
Suspension of disbelief has limits. Movie writers seem to rely on mass ignorance to stretch these limits. Nobody would accept a car chase that suddenly goes airborne because everybody knows that cars can't fly under their own power. That said, some things can be accepted as a visually appealing metaphor, such as banging at the keyboard to hack a computer.
Others, like being knocked out for half a day and actually waking up (and without brain damage no less), simply shows the writer knows nothing about the subject he's writing on. Once you realize the world doesn't follow the same laws as your own, and there is no predictable set of laws that it does follow, then most people lose interest. The semi-predictability is crucial. It's akin to the difference between a conversation and a random series of words. Or a protagonist which has 30 superpowers and solves every conflict by revealing a new one (deus ex machina).